-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hello,

On Sep 1, 2007, at 00:44, khym@azeotrope.org wrote:

> On Mon, Jul 23, 2007 at 12:56:42AM -0400, Michael Lorenz wrote:
>> Anyway, apparently radeonfb fails to program PLLs right on this radeon
>> model, and what it produces is probably pretty close to the monitor's
>> tolerance so it might oscillate between trying to display this signal
>> and ignoring it. Does the monitor go into power saving mode when it
>> blanks out? ( you should be able to hear the difference between
>> blanking and actually turning the monitor off )
>
> I'm not sure... I don't hear any difference, but it's a LCD screen,
> and I haven't been able to tell the difference between blank and 
> powered
> off on those (I can usually tell for a CRT though)

I just committed something that might help in your case ( update 
macppc/macppc/autoconf.c and dev/pci/radeonfb.c ) - the reference clock 
stated by your firmware is what radeonfb would use as default, namely 
27MHz. It does however hardcode the reference divider when it can't get 
data from an x86 BIOS so I had it probe the divider - maybe that's what 
went wrong.

So, please try those changes.

have fun
Michael
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.4 (Darwin)

iQEVAwUBRtsG6spnzkX8Yg2nAQIJDAf/dcrx8QdNxyhIryyg0Br5Zd3okCEUQPQ+
yV5TxZv/Ms+3KrKgQ7rp6WW73fFOkrFk3zJaKSx4l25ZYlp3MYMwUVCwqDBdW75i
hRjyCSV9YMZx80hrOsEzh36mnn3PoDdzWZMeZqbgY87PeErNhx0Y6n0596WMrwZx
GTslJIAP/sa1BhTfrdn9cUro4/m6F600c0UkgIy8stMT3wmq4m6ETzIXTDE33w3S
/GA1mgSu83A0nVANRmkcvdpPuxrUPuAJOlRPijOd7NvtdW+rsQzR0PkHNXwMn6eM
Gc/91CmsarOIDJa2nobB0cK3qVcjftdxBqRivanX+oUhfyLG29wVBA==
=fFcq
-----END PGP SIGNATURE-----